53 ANGIOSPERMAE MONOCOTYLEDON ES 



they have been dusted with pollen from the same flower. At a low temperature 

 (12-13 C), or after several days' rain, the flowers remain closed, and pseudo-l 

 cleistogamous fertilization takes place. 



Rimpau says that at a temperature of 12-13 C. the glumes do open a little,] 

 but not completely under 16 C. This is not, as Godron asserts, limited to the] 

 early morning hours, but may take place at any hour of the day or even in the 

 evening. The length of anthesis depends on the temperature and the dryness 

 the air : at 23 C. the flowers closed again in 15-20 minutes; at a lower temperature] 

 this took place much more slowly. Rimpau found that the anthers dehisce before 

 they have reached the upper margin of the glumes by elongation, so that automatic 

 self-pollination is regularly effected in every flower ; Godron observed the opposite 

 Rimpau and von Liebenberg also found that the species is fertile when cross 

 pollination is excluded. If self-pollination does not take place, crossing is sufficientlj 

 ensured, according to Rimpau's experiments, for out of 85 flowers from which^ 

 the anthers were removed, 50 set fruits. Even in the fourth generation the results 

 of crossing showed a distinct superiority over those produced by autocarpy as regards 

 the average number of haulms formed {cf. Hackel, Bot. Ztg., Leipzig, xxxviii, 1880). 



Kerner states that the anthers dehisce between 5 and 6 a.m. Warnstorf describes 

 the pollen-grains as white in colour, ellipsoidal or ovoid, smooth, opaque, about 

 56 /Lt broad and 75 /^ long. 



3058. T. Spelta L. This species possesses the same flower mechanism as 

 the last one (Hildebrand, Kirchner). Askenasy states that the flowers remain 

 pseudo-cleistogamously closed in unfavourable weather (Hansgirg). 



3059. T. monococcum L. The flower mechanism of this species resembles 

 that of T. vulgare, but the glumes diverge more widely and the whole spike fades 

 more quickly. The glumes diverge in the morning (Kirchner, * Flora v. Stuttgart,' 

 p. 156). Beijerinck also found the flowers to be self-fertile. 



3060. T. dicoccum Schrank. Hildebrand states that in this species the 

 pollination of the stigma is only possible for a few minutes when the glumes diverge 

 slightly, the anthers having then scattered only a part of their pollen. 



3061. T. polonicum L. In this species pseudo-cleistogamous fertilization 

 predominates. In very many flowers the anthers always remain within the glumes 

 and are then apposed to the stigmas. The glumes, however, do diverge for a short 

 time, but only to a distance of about 2 mm. Kornicke says that the anthers usually 

 dehisce at the tip only. 



3062. T. ovatum Rasp. (= Aegilops ovata Z.). Godron says that the flowers 

 of this species, which is indigenous to South Europe, Croatia, &c., open between 

 9.30 and 10 a.m. When the temperature is about 20-1 C. anthesis takes place 

 rapidly, and the anthers tip over before they have dehisced, so that self-pollination is at 

 that time impossible. If the temperature is lower anthesis is so slow that single flowers 

 which have not been fertilized remain open until next day, and can then be dusted 

 with the pollen of flowers belonging to other species, which open 2-3 hours 

 earlier. Godron considers that this explains the fact that at Nancy, which is_ 

 comparatively northerly, such hybrids arise automatically more easily than in moi! 

 southerly districts. (According to Loew, ' Bliitenbiol. Floristik,' p. 372.) Clos 



